PROJECT SUMMARY: Over the past two decades, high rates of opioid prescribing have led to widespread opioid use and epidemic levels of adverse opioid-related outcomes. Early interventions focused on reducing opioid prescribing through the development of practice guidelines and implementation of programs and policies to increase prescription opioid management and monitoring. While overall opioid prescribing has declined in recent years and may have contributed to reductions in prescription opioid-involved overdose death rates in 2012 and 2013, the rate has since risen to a new high, comprising 40% of all opioid-related deaths in 2016. Furthermore, prescribing remains elevated in disadvantaged groups and areas, highlighting a need to better understand how individual characteristics intersect with health system and population-level factors in the pathways from opioid use to fatal and non-fatal overdose. Using a multi-level approach, the proposed project examines trajectories from opioid treatment for chronic non-cancer pain to opioid overdose and fatality, accounting for: 1) individual sociodemographic and clinical characteristics (e.g. sex, age, race/ethnicity, comorbidities), 2) opioid treatment patterns (e.g. dosage, duration, formulation), 3) county-level health system (e.g. facilities, services) and population characteristics (e.g. percent poverty, minority, unemployed, disabled), and 4) state-level policies (e.g. Medicaid policies, Prescription Drug Monitoring Programs). Opioid treatment patterns will be assessed using national Medicaid Analytic Extract (MAX) data, adjusting for individual characteristics. Risk of overdose based on individual characteristics and treatment patterns will be assessed using MAX data for medically treated non-fatal opioid overdose and National Death Index (NDI) data for fatal opioid overdose. Overdose outcomes will be further assessed using county-level Area Health Resource File (AHRF) data and state-level policy measures. Contextualizing individual outcomes within wider health care systems and social and geographic settings supports NIDA?s strategic plan to increase the public health impact of research. Additional substantive and methodological training will contribute to the applicant?s long-term goal to develop an independent academic research career and generate and disseminate robust evidence to inform effective interventions to reduce opioid-related morbidity and mortality. The training aims include advanced study and application of: 1) innovative methods to characterize clinical care patterns over time, 2) social and spatial epidemiology theory and methods to study multilevel influences on health, 3) policy drivers of health and health care, and 4) professional development for conducting ethical, independent research. The research and training are supported by an interdisciplinary mentorship team of Columbia University researchers with expertise in substance use and policy research, clinical and spatial epidemiology, and statistics with additional support from external collaborators in health services and policy research methods. This project and career development plan will prepare the applicant to secure R01 funding to extend this line of research.